BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260625T133338Z
LOCATION:Bldg. 6 - 001 - Plenary Room
DTSTART;TZID=Europe/Stockholm:20260630T122000
DTEND;TZID=Europe/Stockholm:20260630T122100
UID:submissions.pasc-conference.org_PASC26_sess129_pos104@linklings.com
SUMMARY:P34 - Scaling Linear Algebra: Eigenvalue Solvers and Performance T
 rends on Contemporary HPC Systems
DESCRIPTION:Maria Montagna, Sergio Orlandini, and Fabio Affinito (CINECA)\
 n\nLarge-scale eigenvalue problems are a fundamental component of modern s
 cientific simulations in fields such as materials science, computational c
 hemistry, and theoretical physics, where they often represent a dominant c
 omputational bottleneck. The efficiency and scalability of linear algebra 
 and eigenvalue algorithms are therefore critical to fully exploiting conte
 mporary high-performance computing (HPC) architectures. \nThis work presen
 ts a systematic performance comparison of widely used dense eigensolver li
 braries (ScaLAPACK, cuSOLVER, rocSOLVER, cuSOLVERMp, ELPA) across several 
 HPC platforms, including CPU-only and heterogeneous GPU-accelerated system
 s (Leonardo, Pitagora, LUMI, MareNostrum 5, and Fugaku). The study evaluat
 es performance and scaling behaviour over a broad range of matrix sizes, d
 ata types, and precision levels. \nOur results show that GPU-based cuSOLVE
 R library achieves the best performance for small matrix sizes (up to 10^8
  elements), while cuSOLVERMp library provides better performance for mediu
 m to large matrices on a limited number of nodes. However, ScaLAPACK and E
 LPA show the best strong-scaling behaviour. \nThe reproducible benchmark s
 uite provides a practical reference for selecting and tuning eigensolvers 
 in large-scale scientific applications in various HPC environments.\n\nSes
 sion Chair: Tobias Hodel (University of Bern, Switzerland)\n\n
END:VEVENT
END:VCALENDAR